Social cognition in early

psychosis: a potential target

for early intervention?

Dr Andrew Thompson

Consultant Psychiatrist Sussex Early

Intervention in Psychosis Service and Honorary

Senior Fellow, Orygen Research Centre,

Melbourne

Outline of presentation

What is social cognition?

Social cognition in psychosis

Why is social cognition important in psychosis?

Social cognition in at risk for psychosis groups

Treatment approaches for social cognition deficits in psychosis

Outline of presentation

What is social cognition?

What is social cognition?

Definition: domain of cognition that involves the perception, interpretation and processing of social information (Ostrum, 1984)

Distinct from non-social cognition stimuli typically personally relevant and changes

over time

bi-directionality of the stimulus evaluation of bias as well as deficit

4 particular important domains

Emotion recognition/processing

Theory of Mind (ToM)

Attributional style/bias

Social perception/knowledge

Emotion recognition facial affect

recognition

Facial affect recognition -morphing

Emotion recognition alternative

facial morphing tasks

Theory of Mind Sally Anne Task

Theory of Mind visual jokes

Theory of Mind - Hinting task

Lucy is broke but she wants to go out in the evening. She knows that David has just been paid. She says to him: "I'm flat broke! Things are so expensive these days.

QUESTION: What does Lucy really mean when she says this?

Answer: Lucy means Will you lend me some money David ? OR Will you take me out tonight and pay?

ADD: Lucy goes on to say: "Oh well, I suppose I'll have to miss my night out.

QUESTION: What does Lucy want David to do?

Answer: She wants David to lend her money or offer to take her out and pay.

Emotional states in inanimate objects

Attributional style questionnaires

Social perception/knowledge

Role playing social situations that involve interpretation of social rules/context e.g.

white lies, misunderstandings, body language

Navigating familiar social situations e.g. going to the cinema templates or social scripts

Social cognition in psychosis

Social cognition deficits in all 4 domains in

schizophrenia

1) Emotion perception/recognition - specific emotion recognition deficits (both for facial affective expression and prosody-voices), esp for negative emotions (Edwards et al, 2002; Kohler and Brennan, 2004)

2) ToM - appears to be impaired in schizophrenia. Includes a number of different tasks that may target different levels of complexity (1st and 2nd order ToM) and attribution of emotion to inanimate objects (Sprong et al, 2007; Brune, 2005)

3) Social information processing bias -including causal attribution biases (e.g. Locus of Control (LOC)) (Bentall et al, 1997;Bentall & Kinderman, 1996)

4) Social knowledge/perception - decoding non-verbal social cues (Monti & Fingeret, 1987); recognition of familiar social situations (Corrigan et al, 1992); interpersonal problem solving (Bellack et al, 1994)

Savla et al, 2012

Social cognition deficits also in First

Episode Psychosis (FEP) and are relatively

stable over time

Deficits appear to be present in FEP (Edwards et al, 2001; Pinkham et al, 2007: Addington et al, 2008; Bertrand, 2007)

Deficits appear relatively stable through phase of illness - some suggestion that

deficits are slightly worse in acute illness

than in remission (Addington and Addington, 1998; Pinkham et al, 2007; Kee et al, 2003)

Deficits are distinct from neurocognitive

deficits and negative symptoms

Related but distinct from neurocognition and negative symptoms (Sergi et al, 2007; Allen et al, 2007)

Social cognition and neurocognition contribute unique variance to the prediction of social functioning (Allen et al 2007; Pinkham and Penn 2006)

Neural activation circuitry for all three are relatively independent (Pinkham et al, 2003)

Why is social cognition important in

psychosis?

Neurobiological overlap with area affected

in psychosis and plausible theories of

psychotic symptom formation

Overlap between neural areas involved in social cognition and those implicated in aetiology and maintenance of schizophrenia fronto- temporo limbic circuits (Penn et al, 2008)

Number of plausible theories explaining evolution of psychotic symptoms with regard to social cognition deficits failure to metarepresent (ToM) (Frith & Corcoran, 1996)

externalising/personalising attributional style (Bentall et al, 1994)

Social Cognition strongly related to

Social Functioning in Psychosis

Poor performance on ToM tasks associated with social behavioural abnormalities (Brune, 2005)

Emotion perception - linked to social competence, independent living, community involvement and interpersonal relationships (Mueser et al, 1996; Poole et al, 2000)

Social perception has been strongly linked to social behaviour (Appelo et al, 1992; Penn et al, 2002) and vocation-related social skills (Vauth et al, 2004)

Social cognition is a better predictor of social functioning than neurocognition (Brune, 2005; Penn et al, 1996; Vauth et al, 2004)

Social cognition in at risk for psychosis groups

Social cognition in at risk for

psychosis groups

State or trait factor?

Risk factor for developing a psychotic disorder?

SchizophreniaFirst episode psychosisRisk groups for psychosis e.g. family history/ UHR

Risk factors for psychosis e.g. family history

Risk factors for psychosis e.g. family history

First episode psychosisFirst episode psychosis SchizophreniaSchizophrenia

Are social cognition deficits state or trait phenomena?

state

trait

Social Cognition deficits in groups at high risk

for psychosis

Healthy relatives of people with schizophrenia - some social perceptual

deficits (Toomey et al, 1999; Janssen et al, 2003; Mazza et al 2008)

Schizotypal personality - Poor ToM and affect recognition (Pickup, 2006; Williams et al, 2008)

Ultra High Risk (UHR) as an at

risk for psychosis group?

Ultra High Risk criteria

To meet UHR (CAARMS) criteria the young person must either: 1) present with subthreshold psychotic symptoms, or 2) present with definite psychotic symptoms of low frequency 3) have had a brief psychotic episode of less than 1 week where

symptoms spontaneously remit

4) have a first degree relative with a diagnosed psychotic disorder, or 5) have a diagnosis of schizotypal personality disorder

PLUS Have experienced a significant drop in functioning or sustained low functioning over the past year

Some deficits found in UHR group but not in

all studies

Facial affect recognition deficits (Addington et al, 2008; Amminger et al, 2010) but not found in another study (Pinkham et al, 2007)

External attributional bias (Ang et al, 2010) but not Locus of Control (LOC) (Paruch et al, 2006)

ToM deficits (Chung et al, 2008) but another study failed to find such a difference (Couture et al, 2008)

Social perception deficits (Couture et al, 2008)

Areas of uncertainty in the literature

Studies with relatively small numbers, not controlling for IQ and often concentrating on single domains of social cognition

Some inconsistent results from different research groups especially with respect to ToM

Only one study compared the differential performance in controls/UHR/FEP

None linked deficits to social functioning/symptoms

Aims of the study

To investigate whether individuals at ultra-

high risk (UHR) of developing psychosis and

FEP patients are equally impaired in a number

of measures of social cognition

Compare performance on social cognition

measures to levels of social

functioning/psychopathology and

neurocognition

Methodology -subjects

The 2 patient groups recruited from Orygen Youth Health, Melbourne

FEP clinic (EPPIC) patients experience at least one week of daily psychotic symptoms and have had less than 6 months previous treatment

UHR clinic (PACE) - fulfilling UHR criteria assessed by the Comprehensive Assessment of At Risk Mental States (CAARMS) (Yung et al, 2005)

Non psychiatric control participants

Methodology measures

Social cognition: ToM - Hinting task (Corcoran et al, 1995); Visual jokes task (Corcoran et al, 1997);

Emotional triangles task (Boraston et al, 2007)

Emotion recognition DANVA (Nowicki and Duke, 1994)

Social knowledge/social perception MSCEIT (Mayer et al, 2003); Social Comprehension and Schema Task (Corrigan et al, 1995)

Attributional style - NSIE (Nowicki & Duke, 1974)

Social functioning: SOFAS; Role and social functioning scales (Cornblatt et al, 2007)

Psychopathology: BPRS; SANS; DASS

Neuropsychology: WASI/NART; Letter Number Span (verbal working memory); WMS-III Spatial

Span (visual working memory);Trails A and B

Baseline demographics, IQ, psychopathology and social functioning in

the 3 groups

Controls (n=30) UHR (n=30) FEP (n=40) P value

Age (yrs) 20.0 19.4 20.6 0.23

Females (%) 17 (60.7) 12 (48.0) 14 (36.8) 0.16

Years of education 13.3 12.2 12.7 0.12

IQ (WASI) 103.4 105.7 106.7 0.62

Social Functioning:

SOFAS

Role functioning

Social functioning

84.1

8.3

8.8

64.5

6.7

6.5

54.5

5.7

6.5

Effect sizes for deficits on social

cognition tasks

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

ToM - Hinting task ToM - Visual joke task Emotion recognition -

DANVA*

Social perception -

MSCEIT

Social knowledge - SCST-

R

Effect size UHR compared to controls

Effect size FEP compared to controls

Externalising bias in UHR group

compared to controls

0

2

4

6

8

10

12

14

16

18

20

control UHR

Relationship to symptoms and

functioning?

SC associated with some negative symptoms rated on the PANNS but not positive

symptoms

SC measures correlated to measures of functioning in all groups

stronger relationship in controls and FEP than in UHR

stronger relationship to social functioning than role functioning

Summary of the study results

FEP performed significantly worse than controls on all tasks

UHR intermediate performance to FEP controls but only significantly worse on ToM tasks

Differences remains when controlling for IQ/age/gender and multiple testing in the analysis

Externalising bias found in UHR group compared to controls and this was correlated with paranoid symptoms and negative symptoms

SC performance correlated with some negative symptoms and measures of social functioning

Data from other similar studies

Green et al 2011 Schizophrenia Bulletin

What about as a risk factor for

transition to psychosis?

Kim et al 2011 Schizophrenia Research

Worse ToM in those who develop

psychosis in our study

Treatment approaches for social cognition deficits in psychosis

Treatment of social cognitive deficits in

psychosis - current psychosocial approaches

1) Neurocognitive enhancement programs with additional social component

Cognitive Enhancement Therapy (CET) (Hogarty and Flesher 1999a; Hogarty and Flesher 1999b)

2) Training targeting specific cognitive impairments

E.g. facial affect Training of Affect Recognition (TAR) (Wolwer et al, 2005)

3) Training programs with a specific focus on social cognition

SCET (Social Cognition Enhancement Training) (Choi and Kwon 2006)

SCIT (Social Cognition Interaction Training) (Penn et al, 2007)

Do social cognition psychosocial

interventions work?

Kurtz and Richardson, 2012

Potential neuroprotective effect of social

cognitive interventions?

Correlation between improvement in

social cognition and brain volume

Eack et al, 2010

Social Cognition and Interaction

Training

Three distinct phases:

1) Emotion training which involves focusing on defining emotions, emotion mimicry and understanding paranoia

2) Figuring out situations which involves focusing on distinguishing facts from guesses, jumping to conclusions, understanding bad events and attributional style

3) Integration which involves sessions dedicated to checking out guesses in real life by using patients own examples of past social interactions as well as role play.

SCIT example attributional style

characters

Positive effects of SCIT in social cognition and

social functioning in schizophrenia

Improved performance on social cognition measures - emotion perception, social perception, ToM, and attributional style (Combs et al 2007)

Social functioning improved significantly with SCIT training in comparison to the control group, and independent of change in psychopathology

Now using SCIT as part of normal clinical practice in parts of New York State (Roberts et al 2010)

Facelook is designed to help you connect and share with the people in your life and in social situations.

This group program is for EPPIC clients and aims to:This group program is for EPPIC clients and aims to:This group program is for EPPIC clients and aims to:This group program is for EPPIC clients and aims to:

What will be involved:What will be involved:What will be involved:What will be involved:You will be asked to attend facelook once a week for 10 consecutive weeks.

When: Tuesdays from 11am-1.30pm, beginning 6th October 2009

Where: Residence 21, Orygen Youth Health, Parkville

This group program is part of a research project so you will be asked to attend a research interview before the program starts and again after the 10 weeks.What are the benefits:What are the benefits:What are the benefits:What are the benefits:

You will be paid $50 for attending each of the research interviews (i.e $100).

A catered lunch will be provided at each group session/Cab vouchers will be provided if you cannot make it to Orygen on your own.

You will hopefully gain skills that will help with maintaining friendships/ relationships, meeting new people and getting a job.

facelookSign UpAre you an EPPIC client? If yes, join now!

This group program is a joint initiative of Melbourne Neuropsychiatry Research Centre and Orygen Youth Health Research Centre, The University of Melbourne

What emotions are these people showing?

help people better understand and recognise different emotions

improve skills for interpreting social cues

learn new ways to evaluate the likely cause of peoples actions or events

Pilot SCIT in FEP group at EPPIC

Two SCIT groups 12 patients (5 males, 7 females; Mean age= 21.6)

10 week program Each session 2 hours long with lunch in between

Generally good feedback from participants and only one drop out (due to worsening of psychosis) average attendance 69% (range 55-90%)

feasible intervention in this group

Improvements in social and role

functioning

Improvements in some social

cognition tasks

Future work

UHR

Iongitudinal studies e.g. to investigate role of social cognition in development of psychosis and poor functional outcome

FEP

Longitudinal studies of relationship between SC neurocognition and outcome

Randomised trial of SCIT in FEP combined with vocational intervention whether this enhances the effect of a vocational intervention (IPS)

Oxytocin can improve social cognition in

psychosis and possibly symptoms too?

Conclusions

Social cognition deficits are seen in patients with schizophrenia and FEP

These deficits are linked to poor social and occupational functioning

Certain deficits are also seen in those at risk for psychosis and may represent trait or risk factors

Preliminary data from suggests Theory of Mind may be particularly important in those at clinical high risk for psychosis

Conclusions

Current psychosocial approaches to ameliorating social cognition deficits (such as SCIT) in schizophrenia and FEP are promising and may improve both social cognition and social/occupational functioning

Such approaches are feasible in FEP populations and may be particularly important in FEP where potential gains in functioning may be the greatest

Combining these approaches with biological treatments such as Oxytocin is a promising area of research

Acknowledgements and thanks

SCARMS Research team: Alicia Papas Cali Bartholomeusz Stephen Wood Shona Francey Paul Amminger Barnaby Nelson Hok Pan Yuen Alison Yung

Funding sources:o Neurosciences Research Grant, Pfizer o Royal Melbourne Hospital Home

Lottery Granto NHMRC Early Career Fellowship

SCIPPY research team: Eoin Killackey

Cali Bartholomeusz

Stephen Wood

Kelly Allot

Tina Profitt

Hok Pan Yuen

Kathy Martin

Virginia Lui

Lori Schell

Gina Woodhead

All individuals who participated in the studies